ProgrammableMarketing/metabase.com (opens in a new tab)
PDF

Views

  • Plan
  • Roadmap
  • Research

On this page

  • The brief
  • Start here
  • Idea portfolio
  • Effort vs impact
  • This week · 6
  • This quarter · 10
  • Big bets · 3
  • Wildcards · 3
  • How we can help
Generated Nov 4, 2026
  1. Programmable
  2. Playbooks
  3. Metabase
Metabase logo

growth playbook · Nov 4, 2026

metabase.com (opens in a new tab)

Metabase growth playbook

Diagnosis

Metabase has distribution most BI companies would envy: about 332 million Docker pulls, 50.2k GitHub stars, 100,000+ companies and a huge library of integration, metric and dashboard pages.

Its hiring shows the next step: a small marketing team adding a hands-on programs and operations person, a product marketer, a developer advocate, an events manager and a first head of partnerships, with finance asked to define the funnel.

The gaps sit at the points of decision: duplicate and missing comparison pages against the open-source and AI-first rivals, pricing buyers must add up themselves, an empty events page, and agent and Docker installs nobody can attribute. This playbook fixes those first, then builds developer content, meetups, warehouse and AI partnerships, one shared funnel and a Cloud migration path, with an open AI-accuracy benchmark as the big bet.

Growth Score

8/10

Biggest Opportunity

Turn the huge open-source and agent-install footprint into measured trials and paid plans, starting with the pages where buyers compare Metabase to Lightdash, Hex, Sigma, Preset and the big BI suites.

Biggest Risk

Every BI rival now pitches AI you can trust; if Metabase's verification and semantic-layer message is not shown with proof and on the comparison pages, open source alone will not set it apart.

01

Start here

The three moves to make first. Each one is written up in full below.

  1. 1

    One page per rival: merge the duplicate comparisons, surface the hidden ones, and add Lightdash, Hex, Sigma and Preset

    Check Search Console for which URL ranks for "metabase vs quicksight" and "metabase vs looker studio" today.

    High impact First result in 1 month
  2. 2

    A cost estimator on the pricing page that adds seats, AI tokens, transforms and storage

    Write the formula and check it against three real customer bills.

    High impact First result in 3 weeks
  3. 3

    Turn the empty events page into a "bring Metabase to my city" sign-up

    Write the new empty-state copy and the two-question form.

    Medium impact First result in 1 month

02

Your idea portfolio

22 ideas, grouped by when to run them. Open any idea for how to run it: the steps, who to approach, copy you can use, and how to measure it.

  • 9 ideas are marked Unexpected: plays you are unlikely to have tried.

Every idea placed by the effort it takes against the impact it could have. Quick wins sit top left; select a number to open that idea. The filters below narrow the list by status, timing, category and what Programmable can build.

Effort vs impactEach number is an idea. Select one to open it.
Impact →
Quick wins · 3Big bets · 11Fill-ins · 6Reconsider · 2
Effort →

Showing 22 of 22 ideas

This week 6

Quick wins with a clear first action.

  • 01

    Metabase has two pages chasing the same "Metabase vs" searches for Looker Studio and QuickSight (a /lp page and a newer /articles page, each canonical to itself, neither linking to the other), two /lp comparisons (Domo and QuickSight) that are missing from the footer list, and no comparison page at all for the open-source and AI-first rivals that now pitch the same message: Lightdash ("open-source BI ... define metrics once"), Preset (open-source Superset), Hex and Sigma. The new Marketing Programs & Operations hire is asked to publish pages through GitHub, spin up landing pages and A/B test them; this is a ready first project.

    High impactLow effort First result in 1 month Marketing Programs & Operatio…

    First action

    Check Search Console for which URL ranks for "metabase vs quicksight" and "metabase vs looker studio" today.

    How to measure it

    Comparison-page trial starts

    Now
    –
    Target
    +30%
    By
    90 days

    Steps

    1. 1

      Map the overlap

      Search Console queries for each /lp and /articles pair.

      Marketing Programs & Operations 1 day

    2. 2

      Assign roles

      Buying page vs technical deep dive, or merge and redirect.

      Product Marketer 1 day

    3. 3

      Cross-link and retitle

      Distinct titles; links both ways.

      Marketing Programs & Operations 1 day

    4. 4

      Fix the footer

      Add Domo and QuickSight.

      Marketing Programs & Operations 1 hour

    5. 5

      Write four new pages

      Lightdash, Hex, Sigma, Preset from the template, facts checked against each rival's site.

      Product Marketer 2 weeks

    6. 6

      A/B test

      One headline or call-to-action test per page.

      Marketing Programs & Operations 3 weeks

    7. 7

      Report

      Trials per page; keep, rewrite or merge.

      Marketing Programs & Operations monthly

    First result
    1 month
    Cost
    Staff time only
    Owner
    Marketing Programs & Operations

    Smallest test

    Cross-link the two QuickSight pages and add both missing comparisons to the footer today.

    Why it works

    Pick one owner page per rival. For Looker Studio and QuickSight, keep the /lp page as the buying page and turn the article into its technical deep dive (BigQuery, Redshift), linked both ways, with a distinct title and query focus, or merge and redirect if they keep competing. Add Domo and QuickSight to the footer. Then write four new pages from the existing template: Lightdash, Hex, Sigma and Preset, each with an honest "when they are a better fit" section like the Looker Studio page has. Test one variable per page (headline or primary call to action) through GitHub.

    Comparison searches come from people already choosing a tool, and two pages splitting one query usually rank worse than one. The rivals added here are the ones selling "open source", "semantic layer" and "AI you can trust", exactly Metabase's new homepage message.

    Ready to use

    ChecklistComparison page clean-up list

    [ ] One owner URL per rival query
    [ ] /lp <-> /articles links for Looker Studio and QuickSight
    [ ] Distinct titles: buying page vs technical deep dive
    [ ] Domo and QuickSight in the footer list
    [ ] New: Lightdash, Hex, Sigma, Preset
    [ ] "When they are a better fit" on every page
    [ ] One A/B test per page, logged in GitHub

    Risks and how to handle them

    • Redirects lose some long-tail traffic

      Merge only where both pages rank for the same queries; otherwise differentiate.

    • Rival claims go stale

      Date each page and review quarterly.

    Evidence

    • Your websiteComparison pages (checked November 4, 2026): /lp pages exist for Tableau, Looker, Looker Studio, Power BI, Superset, Omni, Domo and QuickSight, but the footer's "Comparisons" list shows only six (Domo and QuickSight are missing). Newer /articles pages target the same rivals: "Metabase vs Google Data Studio (Looker Studio) for BigQuery and Cloud SQL" (dated October 13, 2026) and "Metabase vs Amazon Quick Sight (QuickSight) for Redshift and RDS". Each page has its own canonical tag, both versions are in the sitemap, and the /lp pages do not link to the newer articles.Source
    • Your websiteThe "Metabase alternatives" post compares "Tableau, Power BI, Looker, Cognos, Hex, Lightdash, Omni, and Superset", but there is no dedicated comparison page for Lightdash, Hex, Sigma, Preset or Evidence.Source
    • Competitor researchRivals' own homepages: Lightdash, "the AI-first, open-source BI platform for modern data teams. Connect to dbt, define metrics once"; Preset, "AI-native business intelligence built on Apache Superset ... all open source"; Hex, "The Analytics platform for humans and agents"; Sigma, "The AI runtime for business"; Omni, "turns your data into a source of truth for AI"; Evidence, "Business Intelligence as Code".Source
    • HiringMarketing Programs & Operations (remote US, posted September 11, 2026, reporting to the VP of Marketing): "Publishing and managing webpages via github; optimizing via A/B tests; spinning up landing pages with minimal direction", "Owning smaller channel experiments end to end, from hypothesis to write-up", building "integrations and automations between our tools using APIs, webhooks, MCPs", knowing "what's quietly costing us money", and "Supporting our community events programs". "We're a small, growing marketing team."Source
    • Your websiteHomepage: "Open source AI analytics you can verify" ("Inspect the query behind every answer"); Metabot, an MCP server and a CLI; "Install with AI" prompts that open in Claude Code, Codex or Cursor and install Open Source "No account needed"; "Trusted by 100,000+ companies"; 50.2k GitHub stars; a wall of quotes from X posts; a table contrasting Metabase with a "Vibe-coded dashboard".Source
  • 02

    Metabase's pricing has a base plan, per-user prices above 5 or 10 users, AI at $3.75 per million tokens after 1 million free, transforms after 1,000 runs, and storage per million rows. A buyer comparing Metabase to a rival has to add those up themselves, while the comparison pages promise "pricing you can see".

    High impactLow effort First result in 3 weeks Product Marketer

    First action

    Write the formula and check it against three real customer bills.

    How to measure it

    Pricing-to-trial conversion

    Now
    –
    Target
    +10%
    By
    60 days

    Steps

    1. 1

      Agree inputs

      Plans, seats and usage extras.

      Product Marketer 1 day

    2. 2

      Set defaults

      Typical usage per plan from billing data.

      Business Operations Manager (Finance & Analytics) 2 days

    3. 3

      Build

      Static page component via GitHub.

      Marketing Programs & Operations 1 week

    4. 4

      Link it

      From pricing and comparison pages.

      Marketing Programs & Operations 1 day

    5. 5

      Hand off to sales

      Prefill the sales form above $20,000 a year.

      Marketing Programs & Operations 1 day

    6. 6

      Test

      A/B against the current page.

      Marketing Programs & Operations 3 weeks

    First result
    3 weeks
    Cost
    Staff time only
    Owner
    Product Marketer

    Smallest test

    Add three worked examples (startup, scale-up, embedded) as a table first.

    Why it works

    Add a small estimator: users, plan, monthly AI questions (converted to tokens), transforms and stored rows; show the monthly and yearly total and the self-hosted alternative. Link it from every comparison page and pass the estimate to the sales form for Enterprise-sized answers.

    Predictable pricing is a selling point against consumption-priced rivals; showing the number turns it into proof.

    Risks and how to handle them

    • Estimates read as quotes

      Label as an estimate and show the inputs.

    Evidence

    • Your websitePricing: Open Source free; Starter $100 a month with 5 users and $6 per extra user; Pro $575 a month with 10 users and $12 per extra user; Enterprise "starts at $20,000 / year"; 14-day trials. Usage extras: Metabase's AI service at "$3.75 per 1m tokens" with 1 million included, transforms with 1,000 runs included then $0.01 a run (advanced $0.02), storage "starts at $2 for 1M rows".Source
    • Your websiteComparison pages (checked November 4, 2026): /lp pages exist for Tableau, Looker, Looker Studio, Power BI, Superset, Omni, Domo and QuickSight, but the footer's "Comparisons" list shows only six (Domo and QuickSight are missing). Newer /articles pages target the same rivals: "Metabase vs Google Data Studio (Looker Studio) for BigQuery and Cloud SQL" (dated October 13, 2026) and "Metabase vs Amazon Quick Sight (QuickSight) for Redshift and RDS". Each page has its own canonical tag, both versions are in the sitemap, and the /lp pages do not link to the newer articles.Source
    • HiringProduct Marketer (remote US, posted June 8, 2026): owns "positioning and messaging for one or more product areas (e.g., embedded analytics, self-serve BI, AI features, enterprise/cloud)", launch strategy, "competitor comparisons, pitch decks, one-pagers, and objection handling guides", "ongoing competitive research", "campaigns, SEO narratives", and "win/loss analysis, interviews, and community feedback".Source
  • 03

    The events page currently says "Oh, it seems like there are no upcoming events", while Metabase is hiring an events manager who must pick target cities from community density and usage data and support community-hosted events. Every visitor to that empty page is a demand signal that is thrown away.

    Medium impactLow effort First result in 1 month Global Community Events Manag…

    First action

    Write the new empty-state copy and the two-question form.

    How to measure it

    Volunteer hosts

    Now
    –
    Target
    25
    By
    60 days

    Steps

    1. 1

      Rewrite the empty state

      Recordings plus the form.

      Marketing Programs & Operations 1 day

    2. 2

      Build the form

      City, role, would host.

      Marketing Programs & Operations 1 day

    3. 3

      Promote

      Community forum, social channels and Slack groups.

      Developer Advocate 1 week

    4. 4

      Rank cities

      Sign-ups plus usage data.

      Global Community Events Manager 2 weeks

    5. 5

      Contact hosts

      Offer the toolkit.

      Global Community Events Manager 1 month

    6. 6

      Report

      Sign-ups to events held.

      Global Community Events Manager monthly

    First result
    1 month
    Cost
    $0
    Owner
    Global Community Events Manager

    Smallest test

    Ship the new copy and form this week.

    Why it works

    Replace the empty state with three things: the best recorded workshops, a short form ("Which city? Would you host?") and a link to a host toolkit. Count sign-ups by city and use them with usage data to choose the first meetups.

    It turns a dead end into a list of hosts and cities, the exact input the events manager needs to start.

    Risks and how to handle them

    • Raises expectations

      Say events are planned by demand and reply to every host.

    Evidence

    • Your websiteEvents page (November 4, 2026): "Oh, it seems like there are no upcoming events"; the page lists recorded workshops such as "Faster analytics with Metabase AI" and "Metabase MCP + CLI", and "Metabase at DataConnect 2026".Source
    • HiringGlobal Community Events Manager (remote US): Metabase has run "our flagship Meetabase in Argentina", "customer-hosted events in Madrid" and "casual bar meetups"; the role will "Identify and prioritize target cities and regions based on community density, product usage data", "Support and scale community-hosted events by providing toolkits", run "executive dinner and VIP programming" and "Define and track KPIs for every event type", including "pipeline influence".Source
    • HiringMarketing Programs & Operations (remote US, posted September 11, 2026, reporting to the VP of Marketing): "Publishing and managing webpages via github; optimizing via A/B tests; spinning up landing pages with minimal direction", "Owning smaller channel experiments end to end, from hypothesis to write-up", building "integrations and automations between our tools using APIs, webhooks, MCPs", knowing "what's quietly costing us money", and "Supporting our community events programs". "We're a small, growing marketing team."Source
  • 04

    The homepage and the Open Source page invite people to paste a prompt into Claude Code, Codex or Cursor to install Metabase with "No account needed". It is a smart channel, but nothing in the prompt tells Metabase it was used, and the agent stops at localhost:3000.

    High impactLow effort First result in 2 weeks Marketing Programs & Operatio…

    First action

    Add a source tag to the homepage prompt URL.

    How to measure it

    Agent-install opt-ins

    Now
    –
    Target
    500 a month
    By
    90 days

    Steps

    1. 1

      Tag variants

      Homepage, OSS page, each agent.

      Marketing Programs & Operations 1 day

    2. 2

      Extend the prompt

      Offer the guide and opt-in list.

      Developer Advocate 1 day

    3. 3

      Track

      Visits to the tagged URLs and guide.

      Marketing Programs & Operations 1 day

    4. 4

      Compare

      Agent vs Docker vs jar routes.

      Business Operations Manager (Finance & Analytics) 2 weeks

    5. 5

      Improve

      Test prompt wording.

      Marketing Programs & Operations monthly

    6. 6

      Report

      Into the funnel review.

      Business Operations Manager (Finance & Analytics) monthly

    First result
    2 weeks
    Cost
    $0
    Owner
    Marketing Programs & Operations

    Smallest test

    Tag one prompt and count visits for two weeks.

    Why it works

    Add a tracked source to the URL in each prompt variant (homepage, Open Source page, each agent), and extend the prompt's last step so the agent offers the getting-started guide and the opt-in community and product update list. Compare installs per variant with the Docker and jar routes.

    An agent-led install is a new top of funnel; without a tag it cannot be counted, improved or credited in the funnel review.

    Risks and how to handle them

    • Prompt gets long

      Keep the opt-in step to one line.

    Evidence

    • Your websiteHomepage: "Open source AI analytics you can verify" ("Inspect the query behind every answer"); Metabot, an MCP server and a CLI; "Install with AI" prompts that open in Claude Code, Codex or Cursor and install Open Source "No account needed"; "Trusted by 100,000+ companies"; 50.2k GitHub stars; a wall of quotes from X posts; a table contrasting Metabase with a "Vibe-coded dashboard".Source
    • Your websiteThe Open Source start page gives one prompt to paste into "Any coding agent that can run terminal commands"; it runs the Docker image or the jar and opens localhost:3000. "Migrate to cloud anytime" appears on pricing.Source
    • HiringMarketing Programs & Operations (remote US, posted September 11, 2026, reporting to the VP of Marketing): "Publishing and managing webpages via github; optimizing via A/B tests; spinning up landing pages with minimal direction", "Owning smaller channel experiments end to end, from hypothesis to write-up", building "integrations and automations between our tools using APIs, webhooks, MCPs", knowing "what's quietly costing us money", and "Supporting our community events programs". "We're a small, growing marketing team."Source
    • HiringBusiness Operations Manager, Finance & Analytics (posted October 16, 2026): projects include "Adapting our ARR definition as our business model evolves", "Evaluating the health of our sales and marketing funnel" and "Finding relationships between product initiatives and expansion or churn". A Technical Product Lead (posted April 21, 2026) will "Talk directly to customers"; "Product is led directly by our CEO."Source
  • 05

    Metabase's Docker images have about 332 million pulls and the CLI and embedding SDK together pass 200,000 weekly npm downloads. The Docker Hub README links to docs, Enterprise and pricing as plain links, so traffic from the busiest distribution channel is hard to see.

    Medium impactLow effort First result in 1 month Marketing Programs & Operatio…

    First action

    Edit the Docker Hub README links.

    How to measure it

    Tracked README visits

    Now
    –
    Target
    3,000 a month
    By
    60 days

    Steps

    1. 1

      Tag links

      Docker Hub and npm READMEs.

      Marketing Programs & Operations 1 hour

    2. 2

      Add the Cloud line

      Link to the migration page.

      Marketing Programs & Operations 1 hour

    3. 3

      Add learning links

      Getting-started videos.

      Developer Advocate 1 hour

    4. 4

      Track

      Visits and trials.

      Marketing Programs & Operations 1 day

    5. 5

      Repeat

      GitHub README and release notes.

      Marketing Programs & Operations 1 week

    6. 6

      Review

      Monthly.

      Business Operations Manager (Finance & Analytics) monthly

    First result
    1 month
    Cost
    $0
    Owner
    Marketing Programs & Operations

    Smallest test

    Tag the Docker Hub links today.

    Why it works

    Add tracked links to the Docker Hub and npm READMEs, a one-line "Want us to run it? Migrate to Metabase Cloud" link to a migration page, and a link to the Developer Advocate's getting-started content. Review traffic monthly.

    These pages are read by people who already chose Metabase; a clear next step costs nothing.

    Risks and how to handle them

    • Developers dislike promotion

      One line, plainly useful.

    Evidence

    • Public registriesDistribution: Docker Hub shows about 332 million pulls across the metabase organisation (278.6 million for metabase/metabase, whose README links to docs, Enterprise and pricing); npm shows @metabase/cli at about 131,000 weekly downloads and @metabase/embedding-sdk-react at about 79,000; the YouTube channel has 10,400 subscribers, 231 videos and 1.34 million views.Source
    • Your websitePricing: Open Source free; Starter $100 a month with 5 users and $6 per extra user; Pro $575 a month with 10 users and $12 per extra user; Enterprise "starts at $20,000 / year"; 14-day trials. Usage extras: Metabase's AI service at "$3.75 per 1m tokens" with 1 million included, transforms with 1,000 runs included then $0.01 a run (advanced $0.02), storage "starts at $2 for 1M rows".Source
    • Your websiteThe Open Source start page gives one prompt to paste into "Any coding agent that can run terminal commands"; it runs the Docker image or the jar and opens localhost:3000. "Migrate to cloud anytime" appears on pricing.Source
  • 06

    Real visitors pass Core Web Vitals, so this is not urgent. But in Google's phone test the homepage text appears after 3.8 seconds and render-blocking requests cost about 1.5 seconds; paid, partner and event traffic often lands cold on mobile.

    Low impactLow effort First result in 2 weeks Marketing Programs & Operatio…

    First action

    Run PageSpeed on the /lp template.

    How to measure it

    Lab LCP

    Now
    –
    Target
    Under 2.5s
    By
    30 days

    Steps

    1. 1

      Test templates

      Home, pricing, /lp.

      Marketing Programs & Operations 1 hour

    2. 2

      Find blockers

      Render-blocking CSS and scripts.

      Marketing Programs & Operations 1 day

    3. 3

      Fix

      Inline critical CSS; defer the rest.

      Frontend engineer 2 days

    4. 4

      Re-test

      Lab and field.

      Marketing Programs & Operations 1 day

    5. 5

      Automate

      Performance check on each publish.

      Marketing Programs & Operations 2 days

    6. 6

      Monitor

      Field data monthly.

      Marketing Programs & Operations monthly

    First result
    2 weeks
    Cost
    Engineering time
    Owner
    Marketing Programs & Operations

    Smallest test

    Fix the /lp template first.

    Why it works

    Inline the critical CSS and defer non-critical styles and scripts on the homepage, pricing and /lp templates, then re-test. Keep it as a check in the GitHub publishing workflow.

    A faster first view helps the landing pages the new hire will A/B test, and keeps the passing grade as pages grow.

    Risks and how to handle them

    • Little visible change

      Measure landing-page conversion, not scores.

    Evidence

    • Google PageSpeedHomepage on mobile scores 71/100 (desktop 96); in the phone test the main text appears after 3.8 seconds, and render-blocking requests cost about 1.5 seconds. Real Chrome visitors pass Core Web Vitals (largest paint 1.4 seconds, interaction 157 ms, layout shift 0).Source
    • HiringMarketing Programs & Operations (remote US, posted September 11, 2026, reporting to the VP of Marketing): "Publishing and managing webpages via github; optimizing via A/B tests; spinning up landing pages with minimal direction", "Owning smaller channel experiments end to end, from hypothesis to write-up", building "integrations and automations between our tools using APIs, webhooks, MCPs", knowing "what's quietly costing us money", and "Supporting our community events programs". "We're a small, growing marketing team."Source

This quarter 10

The core plays for the next three months.

  • 07

    Metabase now sells MCP, a CLI and agent installs, the CLI has about 131,000 weekly npm downloads, and the Developer Advocate is hired to build demo projects and screencasts that people "share and return to". The integration guides already show MCP + CLI routes.

    High impactMedium effort First result in 1 quarter Developer Advocate

    First action

    Pick the first four projects from common forum questions.

    How to measure it

    Episode views

    Now
    –
    Target
    5,000 each
    By
    2 quarters

    Steps

    1. 1

      Pick projects

      From forum and Reddit questions.

      Developer Advocate 1 week

    2. 2

      Build repos

      One per episode.

      Developer Advocate weekly

    3. 3

      Record

      Five-minute video and a short.

      Developer Advocate weekly

    4. 4

      Share

      Where the question was asked.

      Developer Advocate weekly

    5. 5

      Feed back

      Friction to product.

      Technical Product Lead monthly

    6. 6

      Measure

      Views, stars, trials.

      Marketing Programs & Operations monthly

    First result
    1 quarter
    Cost
    Staff time only
    Owner
    Developer Advocate

    Smallest test

    Ship one episode and its repo.

    Why it works

    Publish a weekly short project: an agent installs Metabase, connects a public dataset, builds a dashboard from a published metric and sets an alert. Each episode has a repo, a five-minute video and a short. Post them where the questions are asked (Reddit, Discord, Slack groups) only when they answer a real question.

    Developers trust working examples over claims, and the "verify the query" message is easiest to show in an agent workflow.

    Risks and how to handle them

    • Agents change fast

      Pin versions in each repo.

    Evidence

    • HiringDeveloper Advocate (remote US, posted June 8, 2026): "showing up where developers already spend time" (Slack communities, forums, Reddit, Discord, YouTube, Twitch); tutorials, screencasts and shorts; "Build and maintain demo projects, sample workflows, and integrations"; success is content that becomes "genuine resources people share and return to" and credibility "in places we never had to pay to be".Source
    • Your websiteHomepage: "Open source AI analytics you can verify" ("Inspect the query behind every answer"); Metabot, an MCP server and a CLI; "Install with AI" prompts that open in Claude Code, Codex or Cursor and install Open Source "No account needed"; "Trusted by 100,000+ companies"; 50.2k GitHub stars; a wall of quotes from X posts; a table contrasting Metabase with a "Vibe-coded dashboard".Source
    • Your websiteIntegration guides such as "Ahrefs + Metabase: Build SEO Dashboards" explain an MCP + CLI route and a warehouse route, and say plainly that Metabase "does not ship a native Ahrefs connector".Source
    • Public registriesDistribution: Docker Hub shows about 332 million pulls across the metabase organisation (278.6 million for metabase/metabase, whose README links to docs, Enterprise and pricing); npm shows @metabase/cli at about 131,000 weekly downloads and @metabase/embedding-sdk-react at about 79,000; the YouTube channel has 10,400 subscribers, 231 videos and 1.34 million views.Source
  • 08

    Metabase has run Meetabase in Argentina, customer events in Madrid and bar meetups, and wants community-hosted events at scale, with cities picked by community density and product usage.

    High impactMedium effort First result in 1 quarter Global Community Events Manag…

    First action

    Rank the top 20 cities from usage data.

    How to measure it

    Host-led meetups

    Now
    –
    Target
    10 a quarter
    By
    2 quarters

    Steps

    1. 1

      Rank cities

      Usage plus sign-ups.

      Global Community Events Manager 2 weeks

    2. 2

      Write the kit

      Format, slides, demo.

      Global Community Events Manager 2 weeks

    3. 3

      Recruit hosts

      From sign-ups and customers.

      Global Community Events Manager 1 month

    4. 4

      Support

      Swag and promotion.

      Marketing Programs & Operations ongoing

    5. 5

      Capture

      Attendees into the CRM with consent.

      Marketing Programs & Operations per event

    6. 6

      Measure

      Attendance and pipeline influence.

      Global Community Events Manager monthly

    First result
    1 quarter
    Cost
    Swag and venue support
    Owner
    Global Community Events Manager

    Smallest test

    Run one host-led meetup with the kit.

    Why it works

    Rank cities by active instances and Cloud accounts per country, sign-ups from the events form and team travel. Write a kit: format, talk template, sample dashboard demo, swag budget and a sign-in sheet tied to the CRM. Launch ten host-led meetups in a quarter and measure pipeline influence.

    Hosts bring local credibility, and a kit keeps quality even while the team is small.

    Risks and how to handle them

    • Uneven quality

      Kit plus a short host call before each event.

    Evidence

    • HiringGlobal Community Events Manager (remote US): Metabase has run "our flagship Meetabase in Argentina", "customer-hosted events in Madrid" and "casual bar meetups"; the role will "Identify and prioritize target cities and regions based on community density, product usage data", "Support and scale community-hosted events by providing toolkits", run "executive dinner and VIP programming" and "Define and track KPIs for every event type", including "pipeline influence".Source
    • Your websiteEvents page (November 4, 2026): "Oh, it seems like there are no upcoming events"; the page lists recorded workshops such as "Faster analytics with Metabase AI" and "Metabase MCP + CLI", and "Metabase at DataConnect 2026".Source
    • Public filingsUS Form 5500 benefit plan filings show 45 active participants in 2025, up from 37 in 2024 (an estimate of US headcount, not the total team).Source
  • 09

    The first Head of Partnerships is hired to turn AI platform and warehouse relationships into pipeline and to win over the sellers and solutions engineers who influence deals. Metabase's data-source pages are guides; partner sellers need something they can send.

    High impactHigh effort First result in 2 quarters Head of Partnerships

    First action

    Pull the share of paying customers on each warehouse.

    How to measure it

    Partner-sourced pipeline

    Now
    –
    Target
    15% of new
    By
    3 quarters

    Steps

    1. 1

      Rank partners

      By customer share.

      Head of Partnerships 1 week

    2. 2

      Build pages

      One per warehouse.

      Product Marketer 1 month

    3. 3

      Marketplace listings

      Where missing.

      Head of Partnerships 1 quarter

    4. 4

      Enable SEs

      Demo and referral form.

      Head of Partnerships 1 month

    5. 5

      Co-market

      Joint webinar each.

      Global Community Events Manager quarterly

    6. 6

      Report

      Sourced and influenced pipeline.

      Business Operations Manager (Finance & Analytics) monthly

    First result
    2 quarters
    Cost
    Marketplace fees
    Owner
    Head of Partnerships

    Smallest test

    Run one joint webinar with one warehouse.

    Why it works

    For the top three warehouses, publish a partner page with a one-click trial, a joint reference customer and an architecture diagram; list Metabase in their marketplaces where it is not already; give partner solutions engineers a 20-minute demo and a referral form. Track partner-sourced and partner-influenced pipeline.

    Warehouse sellers are in the room when BI is chosen; making Metabase easy for them to recommend is cheaper than outbound.

    Risks and how to handle them

    • Partners favour bigger vendors

      Lead with joint customers and open source reach.

    Evidence

    • HiringHead of Partnerships (posted July 13, 2026), the first in the role: relationships with "the AI platforms and data warehouse vendors our customers build on every day", "co-selling and co-marketing motions that turn partner relationships into pipeline, not just goodwill", and trust with "individual sellers and solutions engineers who influence deals in the field".Source
    • Your websiteMarketing sitemap (1,476 URLs, November 4, 2026): 263 integration guides, 177 dashboard examples, 167 glossary terms, 155 metric pages, 147 blog posts, 132 Learn pages, 45 case studies, 45 partner (Metabase Experts) pages, 53 event pages, 22 analytics use-case pages, a 13-page startup guide and Data Stack Reports for 2023 and 2025.Source
    • Your websiteCase studies: 45 stories, including N26, Gojek, Cal.com, Dribbble, Reforge, Cypress and AngelList; the list page is titled "Real companies, real data, real stories".Source
  • 10

    Metabase sells an MCP server and the homepage shows Claude Code, Codex and Cursor; the partnerships role names "major AI vendors" as the near-term focus. Assistants and agent tools now have connector directories that users browse.

    High impactMedium effort First result in 1 quarter Head of Partnerships

    First action

    List the connector directories and their requirements.

    How to measure it

    MCP server connections

    Now
    –
    Target
    +50%
    By
    2 quarters

    Steps

    1. 1

      List directories

      Requirements for each.

      Head of Partnerships 1 week

    2. 2

      Prepare listings

      Docs, icon, sample prompt.

      Developer Advocate 2 weeks

    3. 3

      Submit

      Each directory.

      Head of Partnerships 1 month

    4. 4

      Co-write

      Tutorial with one vendor.

      Developer Advocate 1 month

    5. 5

      Track

      Installs by source.

      Marketing Programs & Operations ongoing

    6. 6

      Review

      Quarterly.

      Head of Partnerships quarterly

    First result
    1 quarter
    Cost
    Staff time only
    Owner
    Head of Partnerships

    Smallest test

    Submit to one directory.

    Why it works

    List the MCP server in each major assistant's connector or extension directory, with setup docs and a sample prompt; co-write a tutorial with at least one AI vendor; track installs per directory.

    Directory listings are a durable distribution channel for agent users, and AI vendors want useful, trusted data connectors.

    Risks and how to handle them

    • Review delays

      Submit early and in parallel.

    Evidence

    • HiringHead of Partnerships (posted July 13, 2026), the first in the role: relationships with "the AI platforms and data warehouse vendors our customers build on every day", "co-selling and co-marketing motions that turn partner relationships into pipeline, not just goodwill", and trust with "individual sellers and solutions engineers who influence deals in the field".Source
    • Your websiteHomepage: "Open source AI analytics you can verify" ("Inspect the query behind every answer"); Metabot, an MCP server and a CLI; "Install with AI" prompts that open in Claude Code, Codex or Cursor and install Open Source "No account needed"; "Trusted by 100,000+ companies"; 50.2k GitHub stars; a wall of quotes from X posts; a table contrasting Metabase with a "Vibe-coded dashboard".Source
    • HiringDeveloper Advocate (remote US, posted June 8, 2026): "showing up where developers already spend time" (Slack communities, forums, Reddit, Discord, YouTube, Twitch); tutorials, screencasts and shorts; "Build and maintain demo projects, sample workflows, and integrations"; success is content that becomes "genuine resources people share and return to" and credibility "in places we never had to pay to be".Source
  • 11

    The Business Operations Manager is asked to evaluate the health of the sales and marketing funnel and link product initiatives to expansion and churn. Metabase has many entry points (Docker, jar, agent prompt, Cloud trial, embedding SDK), and each team sees its own slice.

    High impactMedium effort First result in 1 quarter Business Operations Manager

    First action

    Write the stage definitions on one page.

    How to measure it

    Weekly funnel review held

    Now
    –
    Target
    12 in a row
    By
    1 quarter

    Steps

    1. 1

      Define stages

      One page of definitions.

      Business Operations Manager (Finance & Analytics) 1 week

    2. 2

      Join sources

      Web, product and billing.

      Business Operations Manager (Finance & Analytics) 1 month

    3. 3

      Publish metrics

      In Data Studio.

      Business Operations Manager (Finance & Analytics) 2 weeks

    4. 4

      Build dashboard

      Funnel by entry point.

      Business Operations Manager (Finance & Analytics) 1 week

    5. 5

      Review weekly

      Marketing, sales, product.

      VP of Marketing weekly

    6. 6

      Share a template

      Public example dashboard.

      Developer Advocate 1 month

    First result
    1 quarter
    Cost
    Staff time only
    Owner
    Business Operations Manager (Finance & Analytics)

    Smallest test

    Build the trial-to-paid stage first.

    Why it works

    Agree definitions for each stage (install or trial, activated, invited teammates, paid, expanded), build the funnel as a Metabase dashboard on published metrics in Data Studio, and review it weekly with marketing, sales and product. Publish a sanitised version as a template.

    Shared definitions stop arguments about numbers, and the funnel itself becomes a showcase of the semantic-layer pitch.

    Risks and how to handle them

    • Self-hosted usage is hard to see

      Use opt-in signals and Cloud data.

    Evidence

    • HiringBusiness Operations Manager, Finance & Analytics (posted October 16, 2026): projects include "Adapting our ARR definition as our business model evolves", "Evaluating the health of our sales and marketing funnel" and "Finding relationships between product initiatives and expansion or churn". A Technical Product Lead (posted April 21, 2026) will "Talk directly to customers"; "Product is led directly by our CEO."Source
    • Your websiteHomepage: "Open source AI analytics you can verify" ("Inspect the query behind every answer"); Metabot, an MCP server and a CLI; "Install with AI" prompts that open in Claude Code, Codex or Cursor and install Open Source "No account needed"; "Trusted by 100,000+ companies"; 50.2k GitHub stars; a wall of quotes from X posts; a table contrasting Metabase with a "Vibe-coded dashboard".Source
    • Public registriesDistribution: Docker Hub shows about 332 million pulls across the metabase organisation (278.6 million for metabase/metabase, whose README links to docs, Enterprise and pricing); npm shows @metabase/cli at about 131,000 weekly downloads and @metabase/embedding-sdk-react at about 79,000; the YouTube channel has 10,400 subscribers, 231 videos and 1.34 million views.Source
    • Your websiteThe Open Source start page gives one prompt to paste into "Any coding agent that can run terminal commands"; it runs the Docker image or the jar and opens localhost:3000. "Migrate to cloud anytime" appears on pricing.Source
  • 12

    Pricing says "Migrate to cloud anytime" and Cloud handles "setup, backups, and upgrades". Self-hosters meet those exact chores, but the moment they feel the pain is inside their instance or the docs, not on the pricing page.

    High impactHigh effort First result in 2 quarters Technical Product Lead

    First action

    List the top docs pages self-hosters visit before migrating.

    How to measure it

    Migrations

    Now
    –
    Target
    +25%
    By
    2 quarters

    Steps

    1. 1

      Find triggers

      Docs and support data.

      Technical Product Lead 2 weeks

    2. 2

      Write the checklist

      Steps and timing.

      Success Engineer 2 weeks

    3. 3

      Place links

      Docs pages at each trigger.

      Marketing Programs & Operations 1 week

    4. 4

      Offer help

      Call for larger instances.

      Success Engineer ongoing

    5. 5

      Measure

      Migrations by trigger.

      Business Operations Manager (Finance & Analytics) monthly

    6. 6

      Improve

      Reduce migration steps.

      Technical Product Lead 1 quarter

    First result
    2 quarters
    Cost
    Staff time only
    Owner
    Technical Product Lead

    Smallest test

    Add the link to the upgrade docs page.

    Why it works

    Find the moments (an upgrade guide visit, a backup doc, an SSO or permissions question) and put a short "Metabase can run this for you" link there, with a migration checklist and a Success Engineer call for larger instances. Measure migrations by trigger.

    Converting existing users is cheaper than finding new ones, and those users already know the product.

    Risks and how to handle them

    • Open-source community resents upsell

      Keep it a quiet, helpful link.

    Evidence

    • Your websitePricing: Open Source free; Starter $100 a month with 5 users and $6 per extra user; Pro $575 a month with 10 users and $12 per extra user; Enterprise "starts at $20,000 / year"; 14-day trials. Usage extras: Metabase's AI service at "$3.75 per 1m tokens" with 1 million included, transforms with 1,000 runs included then $0.01 a run (advanced $0.02), storage "starts at $2 for 1M rows".Source
    • Your websiteThe Open Source start page gives one prompt to paste into "Any coding agent that can run terminal commands"; it runs the Docker image or the jar and opens localhost:3000. "Migrate to cloud anytime" appears on pricing.Source
    • Public registriesDistribution: Docker Hub shows about 332 million pulls across the metabase organisation (278.6 million for metabase/metabase, whose README links to docs, Enterprise and pricing); npm shows @metabase/cli at about 131,000 weekly downloads and @metabase/embedding-sdk-react at about 79,000; the YouTube channel has 10,400 subscribers, 231 videos and 1.34 million views.Source
    • HiringBusiness Operations Manager, Finance & Analytics (posted October 16, 2026): projects include "Adapting our ARR definition as our business model evolves", "Evaluating the health of our sales and marketing funnel" and "Finding relationships between product initiatives and expansion or churn". A Technical Product Lead (posted April 21, 2026) will "Talk directly to customers"; "Product is led directly by our CEO."Source
  • 13

    There are 45 case studies with strong names (N26, Gojek, Cal.com, Dribbble), but the product marketer is asked to own positioning by area: embedded analytics, self-serve BI, AI features and enterprise. Buyers on a comparison page want the story of someone who switched from that rival.

    High impactMedium effort First result in 1 quarter Product Marketer

    First action

    Tag the 45 stories by use case and tool replaced.

    How to measure it

    Story-linked trials

    Now
    –
    Target
    +15%
    By
    2 quarters

    Steps

    1. 1

      Tag stories

      Use case, plan, replaced tool.

      Product Marketer 1 week

    2. 2

      Place them

      On comparison and product pages.

      Marketing Programs & Operations 1 week

    3. 3

      Interview

      Embedded and AI customers.

      Product Marketer 1 month

    4. 4

      Write

      Five new stories.

      Product Marketer 1 quarter

    5. 5

      Cut clips

      Short video quotes.

      Developer Advocate 1 month

    6. 6

      Measure

      Trials from story pages.

      Marketing Programs & Operations monthly

    First result
    1 quarter
    Cost
    Staff time only
    Owner
    Product Marketer

    Smallest test

    Add one "switched from" story to one comparison page.

    Why it works

    Tag every story by use case, plan and the tool replaced; add "switched from" stories to each comparison page; write three new ones for embedded analytics with the React SDK and two for Metabot.

    Relevant proof at the point of decision converts better than a logo wall.

    Risks and how to handle them

    • Customers decline naming

      Offer anonymised stories with numbers.

    Evidence

    • Your websiteCase studies: 45 stories, including N26, Gojek, Cal.com, Dribbble, Reforge, Cypress and AngelList; the list page is titled "Real companies, real data, real stories".Source
    • HiringProduct Marketer (remote US, posted June 8, 2026): owns "positioning and messaging for one or more product areas (e.g., embedded analytics, self-serve BI, AI features, enterprise/cloud)", launch strategy, "competitor comparisons, pitch decks, one-pagers, and objection handling guides", "ongoing competitive research", "campaigns, SEO narratives", and "win/loss analysis, interviews, and community feedback".Source
    • Your websiteComparison pages (checked November 4, 2026): /lp pages exist for Tableau, Looker, Looker Studio, Power BI, Superset, Omni, Domo and QuickSight, but the footer's "Comparisons" list shows only six (Domo and QuickSight are missing). Newer /articles pages target the same rivals: "Metabase vs Google Data Studio (Looker Studio) for BigQuery and Cloud SQL" (dated October 13, 2026) and "Metabase vs Amazon Quick Sight (QuickSight) for Redshift and RDS". Each page has its own canonical tag, both versions are in the sitemap, and the /lp pages do not link to the newer articles.Source
  • 14

    Metabase ships often (Metabase 65 brought data apps and AI Research) and the product marketer will own launches. Each release is a reason to reach users who stopped paying attention.

    Medium impactMedium effort First result in 1 quarter Product Marketer

    First action

    Write the kit as a checklist using Metabase 65.

    How to measure it

    Launch-window trials

    Now
    –
    Target
    +20%
    By
    2 quarters

    Steps

    1. 1

      Write the kit

      Checklist and owners.

      Product Marketer 1 week

    2. 2

      Demo

      Two-minute video.

      Developer Advocate per release

    3. 3

      Workshop

      Live session.

      Global Community Events Manager per release

    4. 4

      Email

      Changelog to opt-in users.

      Marketing Programs & Operations per release

    5. 5

      Update comparisons

      Affected pages.

      Product Marketer per release

    6. 6

      Measure

      Two-week window.

      Marketing Programs & Operations per release

    First result
    1 quarter
    Cost
    Staff time only
    Owner
    Product Marketer

    Smallest test

    Run the kit on the next release.

    Why it works

    For each major release, ship the release page, a two-minute demo, a live workshop, a changelog email and updates to every comparison page the release affects. Track trials and upgrades in the two weeks after.

    A repeatable launch makes each release a growth moment instead of a changelog entry.

    Risks and how to handle them

    • Launch fatigue

      Full kit only for major releases.

    Evidence

    • HiringProduct Marketer (remote US, posted June 8, 2026): owns "positioning and messaging for one or more product areas (e.g., embedded analytics, self-serve BI, AI features, enterprise/cloud)", launch strategy, "competitor comparisons, pitch decks, one-pagers, and objection handling guides", "ongoing competitive research", "campaigns, SEO narratives", and "win/loss analysis, interviews, and community feedback".Source
    • Your websiteMarketing sitemap (1,476 URLs, November 4, 2026): 263 integration guides, 177 dashboard examples, 167 glossary terms, 155 metric pages, 147 blog posts, 132 Learn pages, 45 case studies, 45 partner (Metabase Experts) pages, 53 event pages, 22 analytics use-case pages, a 13-page startup guide and Data Stack Reports for 2023 and 2025.Source
    • Your websiteEvents page (November 4, 2026): "Oh, it seems like there are no upcoming events"; the page lists recorded workshops such as "Faster analytics with Metabase AI" and "Metabase MCP + CLI", and "Metabase at DataConnect 2026".Source
  • 15

    Metabase explicitly allows AI crawlers and publishes an llms.txt, so it is already thinking about answer engines. The question is what assistants actually answer for "best open source BI", "Metabase vs Lightdash" or "Metabase pricing".

    High impactMedium effort First result in 1 quarter Marketing Programs & Operatio…

    First action

    Write the 40 questions.

    How to measure it

    Assistant answers naming Metabase

    Now
    –
    Target
    +20 points
    By
    2 quarters

    Steps

    1. 1

      Write questions

      Buyer, developer and pricing.

      Marketing Programs & Operations 2 days

    2. 2

      Baseline

      Each assistant.

      Marketing Programs & Operations 1 week

    3. 3

      Fix facts

      Site pages and llms.txt.

      Product Marketer 2 weeks

    4. 4

      Add pages

      Missing comparisons.

      Product Marketer 1 month

    5. 5

      Re-check

      Monthly.

      Marketing Programs & Operations monthly

    6. 6

      Report

      To marketing.

      Marketing Programs & Operations monthly

    First result
    1 quarter
    Cost
    Staff time only
    Owner
    Marketing Programs & Operations

    Smallest test

    Run ten questions today.

    Why it works

    Track 40 buyer questions across the main assistants each month, record whether Metabase is named, what pricing and features are cited and which pages are linked. Fix wrong or missing facts on the site (pricing, AI on every plan, self-hosting) and add the missing comparison pages from idea 1.

    Buyers increasingly start in an assistant; a monthly check shows where the story is wrong.

    Risks and how to handle them

    • Answers vary run to run

      Run each question three times.

    Evidence

    • Your websiterobots.txt has an "Explicit allow for AI/LLM crawlers (AEO)" section for GPTBot, ClaudeBot, PerplexityBot and Google-Extended, and an llms.txt describes Metabase as "trusted by over 100,000 companies".Source
    • Competitor researchRivals' own homepages: Lightdash, "the AI-first, open-source BI platform for modern data teams. Connect to dbt, define metrics once"; Preset, "AI-native business intelligence built on Apache Superset ... all open source"; Hex, "The Analytics platform for humans and agents"; Sigma, "The AI runtime for business"; Omni, "turns your data into a source of truth for AI"; Evidence, "Business Intelligence as Code".Source
    • Your websitePricing: Open Source free; Starter $100 a month with 5 users and $6 per extra user; Pro $575 a month with 10 users and $12 per extra user; Enterprise "starts at $20,000 / year"; 14-day trials. Usage extras: Metabase's AI service at "$3.75 per 1m tokens" with 1 million included, transforms with 1,000 runs included then $0.01 a run (advanced $0.02), storage "starts at $2 for 1M rows".Source
    • Your websiteThe "Metabase alternatives" post compares "Tableau, Power BI, Looker, Cognos, Hex, Lightdash, Omni, and Superset", but there is no dedicated comparison page for Lightdash, Hex, Sigma, Preset or Evidence.Source
  • 16

    Metabase has 263 integration guides (Ahrefs, Amplitude, Ashby...) that honestly say there is no native connector and explain the MCP + CLI or warehouse route. They attract people with a specific data source and a specific job.

    Medium impactMedium effort First result in 1 quarter Developer Advocate

    First action

    Rank integration guides by traffic.

    How to measure it

    Integration-guide trials

    Now
    –
    Target
    +30%
    By
    2 quarters

    Steps

    1. 1

      Rank guides

      By traffic.

      Marketing Programs & Operations 1 day

    2. 2

      Build samples

      Top 20 sources.

      Developer Advocate 1 month

    3. 3

      Example dashboards

      One per source.

      Developer Advocate 1 month

    4. 4

      Link pages

      Metrics and dashboard examples.

      Marketing Programs & Operations 1 week

    5. 5

      Track

      Trials per guide.

      Marketing Programs & Operations ongoing

    6. 6

      Extend

      Next 50 guides.

      Developer Advocate 1 quarter

    First result
    1 quarter
    Cost
    Staff time only
    Owner
    Developer Advocate

    Smallest test

    Add a sample dashboard to the top guide.

    Why it works

    Add to each guide a downloadable sample dataset and a one-click example dashboard for Cloud trials, link the matching metric and dashboard example pages, and track which guides start trials.

    Showing the finished dashboard turns a how-to read into a trial.

    Risks and how to handle them

    • Sample data looks fake

      Use realistic public data.

    Evidence

    • Your websiteIntegration guides such as "Ahrefs + Metabase: Build SEO Dashboards" explain an MCP + CLI route and a warehouse route, and say plainly that Metabase "does not ship a native Ahrefs connector".Source
    • Your websiteMarketing sitemap (1,476 URLs, November 4, 2026): 263 integration guides, 177 dashboard examples, 167 glossary terms, 155 metric pages, 147 blog posts, 132 Learn pages, 45 case studies, 45 partner (Metabase Experts) pages, 53 event pages, 22 analytics use-case pages, a 13-page startup guide and Data Stack Reports for 2023 and 2025.Source
    • HiringDeveloper Advocate (remote US, posted June 8, 2026): "showing up where developers already spend time" (Slack communities, forums, Reddit, Discord, YouTube, Twitch); tutorials, screencasts and shorts; "Build and maintain demo projects, sample workflows, and integrations"; success is content that becomes "genuine resources people share and return to" and credibility "in places we never had to pay to be".Source

Big bets 3

Larger plays that can change the trajectory.

  • 17

    Metabase's homepage says "Open source AI analytics you can verify" and argues that a semantic layer guides agents to correct answers, against hand-rolled SQL. Rivals (Lightdash, Omni, Hex, Sigma) make similar AI-trust claims. A public, reproducible test would turn the claim into evidence.

    High impactHigh effort First result in 3 quarters Developer Advocate

    First action

    Draft 50 questions on one public dataset.

    How to measure it

    Repo stars

    Now
    –
    Target
    2,000
    By
    3 quarters

    Steps

    1. 1

      Pick datasets

      Public and realistic.

      Developer Advocate 2 weeks

    2. 2

      Write questions

      200 with known answers.

      Business Operations Manager (Finance & Analytics) 1 month

    3. 3

      Run tests

      With and without metrics.

      Developer Advocate 1 month

    4. 4

      Publish

      Repo, method and results.

      Developer Advocate 1 month

    5. 5

      Launch

      Blog, video and talk.

      Product Marketer 1 month

    6. 6

      Update

      Each major release.

      Developer Advocate quarterly

    First result
    3 quarters
    Cost
    Staff and model costs
    Owner
    Developer Advocate

    Smallest test

    Run 50 questions on one dataset.

    Why it works

    Build an open benchmark on public datasets: 200 business questions with known answers, run by agents with raw schema access and with published metrics, scored for accuracy. Publish the repo, results and method, and invite other tools to run it.

    Open, reproducible evidence earns developer trust, links and press in a way claims cannot.

    Risks and how to handle them

    • Results look self-serving

      Open method; invite others to run it.

    Evidence

    • Your websiteHomepage: "Open source AI analytics you can verify" ("Inspect the query behind every answer"); Metabot, an MCP server and a CLI; "Install with AI" prompts that open in Claude Code, Codex or Cursor and install Open Source "No account needed"; "Trusted by 100,000+ companies"; 50.2k GitHub stars; a wall of quotes from X posts; a table contrasting Metabase with a "Vibe-coded dashboard".Source
    • Competitor researchRivals' own homepages: Lightdash, "the AI-first, open-source BI platform for modern data teams. Connect to dbt, define metrics once"; Preset, "AI-native business intelligence built on Apache Superset ... all open source"; Hex, "The Analytics platform for humans and agents"; Sigma, "The AI runtime for business"; Omni, "turns your data into a source of truth for AI"; Evidence, "Business Intelligence as Code".Source
    • HiringDeveloper Advocate (remote US, posted June 8, 2026): "showing up where developers already spend time" (Slack communities, forums, Reddit, Discord, YouTube, Twitch); tutorials, screencasts and shorts; "Build and maintain demo projects, sample workflows, and integrations"; success is content that becomes "genuine resources people share and return to" and credibility "in places we never had to pay to be".Source
  • 18

    Meetabase in Argentina is the flagship, and the events manager will own the annual conference and executive dinners with pipeline-influence KPIs.

    High impactHigh effort First result in 4 quarters Global Community Events Manag…

    First action

    Shortlist three regions with usage data.

    How to measure it

    Influenced pipeline

    Now
    –
    Target
    Set before launch
    By
    4 quarters

    Steps

    1. 1

      Pick region

      Usage and sign-ups.

      Global Community Events Manager 1 month

    2. 2

      Book venue

      And dates.

      Global Community Events Manager 1 month

    3. 3

      Speakers

      Customers and team.

      Global Community Events Manager 1 quarter

    4. 4

      Promote

      Community and partners.

      Marketing Programs & Operations 1 quarter

    5. 5

      Run

      Event and dinner.

      Global Community Events Manager 1 day

    6. 6

      Measure

      Against goals.

      Business Operations Manager (Finance & Analytics) 1 month

    First result
    4 quarters
    Cost
    Event budget
    Owner
    Global Community Events Manager

    Smallest test

    Run one executive dinner in the shortlisted city.

    Why it works

    Pick the next region by usage and sign-ups, run a one-day Meetabase with customer talks, a builders track and an executive dinner, and set goals before the event (attendees, opportunities influenced, hosts recruited).

    A conference creates content, case studies and relationships for the year.

    Risks and how to handle them

    • Cost per lead high

      Reuse content for months after.

    Evidence

    • HiringGlobal Community Events Manager (remote US): Metabase has run "our flagship Meetabase in Argentina", "customer-hosted events in Madrid" and "casual bar meetups"; the role will "Identify and prioritize target cities and regions based on community density, product usage data", "Support and scale community-hosted events by providing toolkits", run "executive dinner and VIP programming" and "Define and track KPIs for every event type", including "pipeline influence".Source
    • Your websiteEvents page (November 4, 2026): "Oh, it seems like there are no upcoming events"; the page lists recorded workshops such as "Faster analytics with Metabase AI" and "Metabase MCP + CLI", and "Metabase at DataConnect 2026".Source
    • Your websiteCase studies: 45 stories, including N26, Gojek, Cal.com, Dribbble, Reforge, Cypress and AngelList; the list page is titled "Real companies, real data, real stories".Source
  • 19

    Metabase has published Community Data Stack Reports for 2023 and 2025. With 100,000+ companies using it, it can survey more data people than almost any vendor.

    High impactMedium effort First result in 3 quarters Business Operations Manager

    First action

    Draft the survey with last year's questions.

    How to measure it

    Survey responses

    Now
    –
    Target
    5,000
    By
    3 quarters

    Steps

    1. 1

      Draft survey

      Keep trend questions.

      Business Operations Manager (Finance & Analytics) 2 weeks

    2. 2

      Distribute

      Community, partners, events.

      Marketing Programs & Operations 1 month

    3. 3

      Analyze

      By segment.

      Business Operations Manager (Finance & Analytics) 1 month

    4. 4

      Publish

      Interactive dashboard.

      Developer Advocate 1 month

    5. 5

      Press

      Findings piece.

      Product Marketer 2 weeks

    6. 6

      Partners

      Segment data.

      Head of Partnerships 1 month

    First result
    3 quarters
    Cost
    Survey incentives
    Owner
    Business Operations Manager (Finance & Analytics)

    Smallest test

    Publish one 2025 chart as an interactive dashboard.

    Why it works

    Run the 2026 survey, publish results as an interactive Metabase dashboard with filters by company size and region, release a short findings piece and share segment data with partners and press.

    Original data earns links and AI citations, and the dashboard demonstrates the product.

    Risks and how to handle them

    • Low response

      Offer early access to results.

    Evidence

    • Your websiteMarketing sitemap (1,476 URLs, November 4, 2026): 263 integration guides, 177 dashboard examples, 167 glossary terms, 155 metric pages, 147 blog posts, 132 Learn pages, 45 case studies, 45 partner (Metabase Experts) pages, 53 event pages, 22 analytics use-case pages, a 13-page startup guide and Data Stack Reports for 2023 and 2025.Source
    • Your websiteHomepage: "Open source AI analytics you can verify" ("Inspect the query behind every answer"); Metabot, an MCP server and a CLI; "Install with AI" prompts that open in Claude Code, Codex or Cursor and install Open Source "No account needed"; "Trusted by 100,000+ companies"; 50.2k GitHub stars; a wall of quotes from X posts; a table contrasting Metabase with a "Vibe-coded dashboard".Source
    • HiringHead of Partnerships (posted July 13, 2026), the first in the role: relationships with "the AI platforms and data warehouse vendors our customers build on every day", "co-selling and co-marketing motions that turn partner relationships into pipeline, not just goodwill", and trust with "individual sellers and solutions engineers who influence deals in the field".Source

Wildcards 3

Unexpected experiments, each with a cheap way to test it.

  • 20

    Metabase already has a 13-page startup guide on financial modelling and a metrics store, and its homepage quotes founders recommending it. Startups choose a BI tool early and keep it.

    Medium impactLow effort First result in 1 quarter Marketing Programs & Operatio…

    First action

    List ten accelerators with perk portals.

    How to measure it

    Programme sign-ups

    Now
    –
    Target
    300
    By
    2 quarters

    Steps

    1. 1

      List programmes

      Accelerators and platforms.

      Marketing Programs & Operations 1 week

    2. 2

      Design perk

      Credits and templates.

      VP of Marketing 1 week

    3. 3

      Template pack

      Startup metrics.

      Developer Advocate 2 weeks

    4. 4

      Submit

      To perk portals.

      Marketing Programs & Operations 1 month

    5. 5

      Track

      Sign-ups by programme.

      Marketing Programs & Operations ongoing

    6. 6

      Review

      Retention after a year.

      Business Operations Manager (Finance & Analytics) quarterly

    First result
    1 quarter
    Cost
    Credits
    Owner
    Marketing Programs & Operations

    Smallest test

    Get listed on one perk portal.

    Why it works

    Offer accelerator and VC platform teams a perk: Starter credits, the startup guide and a template dashboard pack. Track sign-ups by programme.

    Accelerator perks reach founders at the moment they set up their stack.

    Risks and how to handle them

    • Low-value accounts

      Credits expire; measure conversion.

    Evidence

    • Your websiteMarketing sitemap (1,476 URLs, November 4, 2026): 263 integration guides, 177 dashboard examples, 167 glossary terms, 155 metric pages, 147 blog posts, 132 Learn pages, 45 case studies, 45 partner (Metabase Experts) pages, 53 event pages, 22 analytics use-case pages, a 13-page startup guide and Data Stack Reports for 2023 and 2025.Source
    • Your websiteHomepage: "Open source AI analytics you can verify" ("Inspect the query behind every answer"); Metabot, an MCP server and a CLI; "Install with AI" prompts that open in Claude Code, Codex or Cursor and install Open Source "No account needed"; "Trusted by 100,000+ companies"; 50.2k GitHub stars; a wall of quotes from X posts; a table contrasting Metabase with a "Vibe-coded dashboard".Source
    • Your websitePricing: Open Source free; Starter $100 a month with 5 users and $6 per extra user; Pro $575 a month with 10 users and $12 per extra user; Enterprise "starts at $20,000 / year"; 14-day trials. Usage extras: Metabase's AI service at "$3.75 per 1m tokens" with 1 million included, transforms with 1,000 runs included then $0.01 a run (advanced $0.02), storage "starts at $2 for 1M rows".Source
  • 21

    Metabase lists 45 partner experts and offers professional services itself. Experts meet companies who need BI help and can bring Cloud and Pro deals.

    Medium impactLow effort First result in 1 quarter Product Marketer

    First action

    Ask the ten most active experts what they need.

    How to measure it

    Referred accounts

    Now
    –
    Target
    50
    By
    2 quarters

    Steps

    1. 1

      Interview experts

      Ten most active.

      Head of Partnerships 2 weeks

    2. 2

      Set rules

      Lead sharing and fees.

      Head of Partnerships 2 weeks

    3. 3

      Referral links

      Tracked.

      Marketing Programs & Operations 1 week

    4. 4

      Badges and pages

      Per expert.

      Marketing Programs & Operations 2 weeks

    5. 5

      Share leads

      Small deals.

      Success Engineer ongoing

    6. 6

      Report

      Quarterly.

      Business Operations Manager (Finance & Analytics) quarterly

    First result
    1 quarter
    Cost
    Referral fees
    Owner
    Product Marketer

    Smallest test

    Run referral links with five experts.

    Why it works

    Give experts a referral link, a partner badge, a co-marketing page per expert and a shared list of leads too small for Enterprise. Report deals referred each quarter.

    Local experts extend reach in regions and languages the small team cannot cover.

    Risks and how to handle them

    • Quality varies

      Badge only for experts with reviews.

    Evidence

    • Your websiteMarketing sitemap (1,476 URLs, November 4, 2026): 263 integration guides, 177 dashboard examples, 167 glossary terms, 155 metric pages, 147 blog posts, 132 Learn pages, 45 case studies, 45 partner (Metabase Experts) pages, 53 event pages, 22 analytics use-case pages, a 13-page startup guide and Data Stack Reports for 2023 and 2025.Source
    • HiringHead of Partnerships (posted July 13, 2026), the first in the role: relationships with "the AI platforms and data warehouse vendors our customers build on every day", "co-selling and co-marketing motions that turn partner relationships into pipeline, not just goodwill", and trust with "individual sellers and solutions engineers who influence deals in the field".Source
  • 22

    The homepage is full of quotes from developers and founders on X, and the Developer Advocate is measured on content people share. A showcase of real dashboards and projects makes those voices regular contributors.

    Medium impactLow effort First result in 1 quarter Developer Advocate

    First action

    Ask five quoted users to share what they built.

    How to measure it

    Monthly submissions

    Now
    –
    Target
    30
    By
    2 quarters

    Steps

    1. 1

      Ask

      Five quoted users.

      Developer Advocate 1 week

    2. 2

      Build page

      Showcase template.

      Marketing Programs & Operations 1 week

    3. 3

      Feature

      Five a month.

      Developer Advocate monthly

    4. 4

      Video

      Short walk-through.

      Developer Advocate monthly

    5. 5

      Badges

      For featured builders.

      Marketing Programs & Operations 1 week

    6. 6

      Measure

      Submissions and shares.

      Marketing Programs & Operations monthly

    First result
    1 quarter
    Cost
    $0
    Owner
    Developer Advocate

    Smallest test

    Feature three builders this month.

    Why it works

    Each month, feature five community dashboards or open repos (with permission) on a showcase page, a short video and the social channels; give featured builders a badge.

    Recognition invites more sharing, which feeds the next month's showcase.

    Risks and how to handle them

    • Private data in dashboards

      Public data or screenshots approved by the builder.

    Evidence

    • Your websiteHomepage: "Open source AI analytics you can verify" ("Inspect the query behind every answer"); Metabot, an MCP server and a CLI; "Install with AI" prompts that open in Claude Code, Codex or Cursor and install Open Source "No account needed"; "Trusted by 100,000+ companies"; 50.2k GitHub stars; a wall of quotes from X posts; a table contrasting Metabase with a "Vibe-coded dashboard".Source
    • HiringDeveloper Advocate (remote US, posted June 8, 2026): "showing up where developers already spend time" (Slack communities, forums, Reddit, Discord, YouTube, Twitch); tutorials, screencasts and shorts; "Build and maintain demo projects, sample workflows, and integrations"; success is content that becomes "genuine resources people share and return to" and credibility "in places we never had to pay to be".Source
    • Public registriesDistribution: Docker Hub shows about 332 million pulls across the metabase organisation (278.6 million for metabase/metabase, whose README links to docs, Enterprise and pricing); npm shows @metabase/cli at about 131,000 weekly downloads and @metabase/embedding-sdk-react at about 79,000; the YouTube channel has 10,400 subscribers, 231 videos and 1.34 million views.Source

03

How Programmable can help

Where an idea in this playbook is something Programmable builds or runs, it's listed here with what we'd do. Everything else is yours to run.

ProgrammableTalk

Fill the events in this plan

These ideas are events. Talk gets registrants for them and tells you who's coming and what their company does before the day.

Plan an event (opens programmable.talk in a new tab)
Website review of metabase.com More example playbooks and reviews
ProgrammableDigital

Show Metabase’s product working

Turn screenshots or the website into an animated product demo in its own brand, for the homepage, launch posts and outreach in this playbook.

Make a demo (opens programmable.digital in a new tab)

More growth playbooks

Other companies we have researched, each read from what they are hiring for.

  • Garner Health growth strategyDoctor quality analytics and incentives: Garner ranks doctors on 550+ metrics from a dataset of 320 million patients, and gives members of…21 growth ideas
  • Integral Molecular growth strategyA Philadelphia biotech, founded in 2001, that discovers and characterises antibodies against membrane proteins: the Membrane Proteome…21 growth ideas
  • TryNearby growth strategyA done-for-you local creator service for restaurants. Every month it sends a new group of vetted nearby food creators to visit, film, edit…22 growth ideas
  • Albemarle growth strategyA global specialty chemicals company producing lithium compounds for batteries (Energy Storage), bromine flame retardants, amines and…21 growth ideas
  • Kepler growth strategyA verifiable AI platform for high-stakes research, starting with finance: the model plans the work while deterministic code retrieves and…21 growth ideas
  • Piermont Bank growth strategyA New York commercial bank, describing itself as the first cloud-native bank launched in the US, offering business and personal banking…21 growth ideas
  • Clara growth strategyAI-powered primary care. Clara imports a member's records from 150,000+ hospitals, labs and pharmacies, chats with them about their…22 growth ideas
  • Elyra growth strategyAn AI reservation and guest platform for restaurants: a voice agent that answers every call, an email agent, real-time table allocation…24 growth ideas

See every playbook and website review

Independent research. Programmable wrote this playbook from public sources, as they stood on November 4, 2026, and cites them throughout. Metabase did not commission, review or endorse it, and Programmable is not affiliated with Metabase. Company and product names are trademarks of their owners, used only to identify them. The ideas are suggestions and opinion, not professional, legal or financial advice, and not statements about how Metabase runs its business.

Something wrong, or do you represent Metabase? Email [email protected] and we will correct it or take it down.

30/60/90 roadmap

What to focus on in each month, and the ideas that carry it.

  1. Days 1 to 306 ideas

    Comparison clean-up and new rival pages, the pricing estimator, the events page sign-up, tagged install prompts and READMEs, and the landing-page speed fix.

  2. Days 31 to 6013 ideas

    Agent project series, meetup kit, warehouse and AI partnerships, the shared funnel, the Cloud migration path, case studies, launch kit, assistant checks and integration guides; start the startup programme, experts channel and showcase.

  3. Days 61 to 903 ideas

    Commit to the big bets: the open AI benchmark, a second Meetabase and the yearly Data Stack Report.

The research behind the plan

3 research areas, each summarised in a few lines. Open any of them for the full detail.

Customers

What customers say

Customers value easy for non-technical people to explore data. The most cited problem: running upgrades and backups themselves.

  • Fast to install and self-host
  • Comparing total cost across seats and usage
  • In their words: “just works”

Market

Competitive landscape

6 competitors compared: Lightdash, Preset (Apache Superset), Hex, Sigma and 2 more.

  • Gap to exploit: Verified, reproducible AI answers
  • Lightdash: Depends on dbt
  • Preset (Apache Superset): Harder for non-technical users

Hiring signals

Metabase's small marketing team is adding hands-on builders: a Marketing Programs & Operations hire who publishes pages through GitHub, A/B tests landing pages and runs channel experiments (ideas 1, 4, 5, 6, 15), a…

  • Marketing Programs & Operations: Landing pages via GitHub, A/B tests and channel experiments. Ideas 1, 4, 5, 6, 15.
  • Product Marketer: Positioning, competitive research and launches. Ideas 1, 2, 13, 14.
  • Developer Advocate: Demo projects and content developers share. Ideas 7, 16, 17, 22.

From the team that wrote this playbook

All products
  • ProgrammableAgencyWorking AI systems in days, not quarters (opens programmable.agency in a new tab)
  • ProgrammableInfoAn agent that finds, researches and asks first (opens programmable.info in a new tab)
  • ProgrammableDigitalScreenshots in, an animated product demo out (opens programmable.digital in a new tab)
  • ProgrammableWatchKnow the day a supplier falls out of compliance (opens programmable.watch in a new tab)
  • ProgrammableTalkGet registrants for your events, and know who’s coming (opens programmable.talk in a new tab)
  • ProgrammablePageText what you need, your customer gets a page that works (opens programmable.page in a new tab)
  • ProgrammableCommunityForums with labelled AI members, never empty (opens programmable.community in a new tab)
ProgrammableMarketing
ExamplesLeaderboardsProductsTermsPrivacyCookiesDPALegalContact

© 2026 AI-Powered Growth Intelligence